Algopix: eCommerce Product Matching Using Deep Learning Technology

Our cutting-edge technology facilitates accurate product matching across 17 global Amazon, eBay and Walmart marketplaces. Algopix provides valuable and reliable market insights into profit margins, selling expenses, demand level and competition – in real time.

Since its beginnings more than 25 years ago, the eCommerce industry has come a long way. Only within recent years, the advancements of online technology, the developments of global internet accessibility and the success stories of eCommerce giants Amazon, eBay, and Walmart have fueled the rise of online business activities on a global scale. This has led to the exponential rise in the number of products being sold across eCommerce stores.

Today’s eCommerce industry allows online merchants to seamlessly offer their products across multiple international online marketplaces with a few clicks. Certainly, that’s great news for all merchants, as it helps to expand their global sales territory, reach a wider customer target group and ultimately, gain higher profits.Consequently, the growing volume of products distributed through online marketplaces has increased the complexity for consumers, vendors, and market research companies, resulting in a need to generate market insights.

Nowadays, identical products may be offered across multiple international eCommerce stores, while product listing attributes such as item description, specifications, classification and identifier are not standardized across marketplaces. Determining product listing matches based on unique identifiers and keywords demands technical finesse and knowledge of the eCommerce industry. In addition, Amazon and eBay product matching both require an individually designed technical approach.

To better understand the problems and complexity that arise with product matching, imagine that a user is looking to find product-market insights on Algopix for the iPhone 8. On our platform, users either research by entering a unique product identifier that corresponds to the iPhone 8 inside the Single and Bulk Product Analysis, or by searching with the “iPhone 8” keyword using our Algopix Product Discovery Tool.

Here’s the tricky part: Identifying the exact product that a user is searching, together with its corresponding live listings and item attributes, is impossible with just a reference to its title or single identifier. All popular eCommerce marketplaces such as Amazon, eBay or Walmart, have been developed and are being maintained by teams that have their own unique approach to presenting a product in their online stores; for instance, an iPhone 8. In some cases, the product title could include a category such as “Smart Phone,” or it could include a specific type of warranty for a product in a particular marketplace. Moreover, product identifiers allocated to a specific item vary among Amazon, eBay and Walmart marketplaces and most often, are not linked with each other.

How does Algopix understand that iPhone 8 64 GB, Fully Unlocked, Red Special Edition and Apple iPhone 8 (PRODUCT)RED – 64GB – (Unlocked) A1778 (GSM) both describe the same product?

Accurate product matching is essential for generating reliable and actionable market insights. We built advanced algorithms and implemented deep learning technology to ensure that Algopix identifies the exact item that a user is researching and to provide real-time actionable product market insights for eCommerce merchants.

Our approach consists of leveraging machine learning technology, based on a set of cutting-edge algorithms, that enriches the product information from the original keyword or identifier search with analytical insights. These insights are generated from multiple product- and market data-points and converted into unique item attributes.